# liyown/ai-trend-publish

TrendPublish: 全自动 AI 内容生成与发布系统 | 微信公众号自动化 | 多源数据抓取 (Twitter/X、网站) | DeepseekAI、千问、讯飞模型 | 智能内容分析排序 | 定时发布 | 多模板支持 | Node.js | TypeScript | AI 技术趋势跟踪工具

Repository: https://github.com/liyown/ai-trend-publish
Canonical: https://ross.abutalabs.com/products/ai-trend-publish
Homepage: https://liyown.github.io/ai-trend-publish/
Language: TypeScript
License: MIT
License Family: permissive
Topics: ai, weixin
Last push: 2026-08-04T14:17:45+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 96, release rhythm 87, longevity 42
- inputs: {"age_days": 597, "days_push": 29, "days_rel": 90, "gap_med": 0, "n_releases_24m": 20}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3158, forks 438 (observed 2026-08-28T04:07:46.462247+00:00)

## What it is
TrendPublish is a TypeScript-based automated content pipeline for WeChat Official Accounts that scrapes multiple sources (Twitter/X, RSS, search APIs, arXiv, Hacker News), uses LLMs for topic selection, evidence gathering, article writing, review, and image generation, then publishes drafts or dry-run artifacts. It includes a dashboard for runtime configuration, run observability, and multi-account matrix operation, deployable locally, via Docker, or on Cloudflare Workers.

## Use cases
- automatically publish daily AI news to a WeChat official account
- generate WeChat articles from Twitter and RSS sources with AI
- automate topic selection and article writing for a tech newsletter
- run a scheduled AI content curation and publishing pipeline
- preview AI-generated WeChat articles with dry-run before publishing
- manage multiple WeChat accounts with automated content generation

## When to choose
- you run a daily-updated WeChat official account and want to automate sourcing, writing, and publishing
- you need an observable, configurable article pipeline with quality review and dry-run support rather than a simple summarizer
- you want multi-source content discovery including Twitter/X, arXiv, and search APIs
- you operate multiple WeChat accounts and need matrix publishing with per-account tone

## When to avoid
- you need publishing to platforms other than WeChat (e.g., Xiaohongshu, email) as a first-class feature
- you want fully hands-off publishing without any human review
- you need a simple RSS-to-email or static-site generator instead
- you cannot provide LLM API keys or deploy on Node.js/Docker/Cloudflare

## Facets
- artifact type: application
- maturity: active
- function: web-scraping, nlp, llm-inference, workflow-automation, scheduling, templating, image-processing, rag, agent-framework
- domain: web-development, artificial-intelligence, social-media, crawlers
- platform: cloud, self-hosted
- tags: wechat, weixin-official-account, content-generation, content-publishing, typescript, deepseek, twitter-scraping, rss, cloudflare-workers, dashboard, dry-run, editorial-automation, automation, natural-language-processing, nodejs, docker, web-server

## Member repositories
- liyown/ai-trend-publish (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:46.462247+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:25:36.464680+00:00, confidence not recorded.
  - readme: https://github.com/liyown/ai-trend-publish (fetched 2026-08-28T04:07:46.462247+00:00, sha 74af0de2b3ee)
  - homepage: https://liyown.github.io/ai-trend-publish/ (fetched 2026-08-29T09:40:11.406565+00:00, sha d6e8970f1d65)
- Data as of 2026-08-30T08:39:29.467469+00:00.